The Value Of Ultrasound-based Radiomics And Nomograph Model In Identifying Benign And Malignant BI-RADS 4a Irregular Breast Masses

Objective: To investigate the value of ultrasound-based radiomics in discriminating benign and malignant BI-RADS category 4a irregular breast nodules,and to combine radiomics,ultrasound features and clinical independent risk factor characteristics to establish a nomogram to assess its value in reducing unnecessary biopsies.Methods: A retrospective collection of 905 irregular breast nodules of BI-RADS 4a pattern screened by conventional ultrasound was randomly divided into a training set(n=634)and a validation set(n=271)in a ratio of 7:3.851 features were extracted using Pyradiomics and a radiomics model was constructed by logistic regression model using surgical pathology as the gold standard.The diagnostic effectiveness of the radiomics model and the nomogram model for irregular breast nodules in ultrasound BI-RADS 4a pattern was assessed by ROC curves.Results:Among 905 irregular breast nodules,485 were malignant nodules and 420 were benign nodules.The patients were divided into training cohort and validation cohort by randomization in the ratio of 7:3,age range22-83 years,mean age about 50.05±11.13 years.The mean age,rad-score value,and mass diameter of the two groups were not significantly different,and the difference was not statistically significant(P>0.05).The AUC value of the training cohort radiomics model was 0.927(95% CI: 0.900-0.950),and the AUC value of the validation cohort radiomics model was 0.946(95% CI:0.908-0.976).The sensitivity,accuracy,specificity,F1 value,and precision for the training cohort of this model were 0.879,0.879,0.877,0.909,and0.940,respectively,and the sensitivity,accuracy,specificity,F1 value,and precision for the validation cohort of this model were 0.890,0.896,0.909,0.921,and 0.956,respectively.The calibration curves showed good calibration between the training and validation cohorts of this mode;The AUC value was 0.943(95% CI: 0.912-0.960)for the training cohort normogram model and 0.968(95% CI: 0.924-0.970)for the validation cohort normogram model,showing the good clinical application of normograms.Conclusion:The ultrasound imaging histology and nomogram models have some helpful value in improving the diagnostic efficacy of benign and malignant BI-RADS class4 a morphologically irregular breast masses,and can provide value to junior physicians in the classification and diagnostic accuracy of BI-RADS class 4a irregular breast masses,which is expected to provide some diagnostic value to clinicians in reducing unnecessary puncture biopsies in the future.